Fusion-io ioDrive 2 1.2TB Reference Page

Louis

New Member
Jul 7, 2019
2
0
1

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38

YardBouncer

always yield to the hands-on imperative
Jul 13, 2019
50
27
18
UK
Hi all,

Excuse my selfishness, long time lurker and now I'm signed up I'm going straight in with a question.

I have an HP Z820 workstation, its the second variant that supports Ivy Bridge. C602 chipset, PCI-e 3 slots.
2x E5-2637 v2 + 64GB. 480GB Revodrive 3x2 as boot drive (no native NVM-e support).
My original plan was to use a HP P812 RAID card for all my other storage but it turns out that secondary controller cards can only go in the top (half length) slot. P812 is absolutely enormous and won't fit.

Consequently I've been looking at Fusion-io cards for bulk fast storage.

I've done a lot of reading; docs for the drivers etc but there's a couple of things I'd like to clarify, if possible.
Being a Starving Hacker spending ~£200 on a card that may not work is quite a gamble.
I've read the thread so far.
I'm fairly competent with computers and pretty good with electronics so can easily make up an external power cable if reqd (although in the photos it looks like this variant doesn't use one).

The main question I have regards drivers etc. I've signed up to the Sandisk site to get the docs.

I'd prefer to stick with Win7 x64 for now although I plan to set up dual boot with 10 just to get to know it.
Do the Server 2008 (R2) drivers work with Win7 x64?

The particular card I'm looking at is a Fusion-io 3.2TB io Scale Accelerator Card F11-002-3T20-CS-0001
I don't think its a branded variant. There's a photo of the back of the card with all it's stickers here:


It seems like there is a lot of experience using these cards here so I hope to be as sure as possible I'm not buying a paperweight before I pay money for this...

Thanks for any help or hints.
 

Attachments

  • Like
Reactions: Samir

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
Yeah, the windows drivers work in all versions of windows for an ioDrive2

For ioDrive 3 (SX/PX series) there are two different executables, one for older server 2008, and the newer executable for all newer windows versions.

-- Dave
 

YardBouncer

always yield to the hands-on imperative
Jul 13, 2019
50
27
18
UK
Thanks very much, I was hoping you'd reply but didn't ask you directly since I bet you get asked about these things constantly.

Useful info on generation too, I wasn't certain which it was. I imagine its all in the model number ie the -002- bit in F11-002-3T20-CS-0001.

I'll get it ordered this week and get fiddling. After I've worked out if I'm the meat or the metal I will make another post with a short description of what I had to do in case it helps anyone else.

Thanks again for the information, I bet you've made it possible for hundreds of these cards to keep going.
Crazy that they are seen as old when they have such enormous capacity and speed compared to SATA. I suppose its the non-bootable thing + complex driver & firmware that puts people off.

One last question re PCI-e slots:

The Z820 has the possibility to switch off option ROM download for particular slots. Since the Revodrive 3x2 uses one and is seen as a RAID card and I'm going to use a HP P420 RAID card for my rust drives I'd like to not have another drive controller option ROM stirring the pot.

Since they are non-bootable I assume they don't use an option ROM? Seems like all the magic happens in the OS; FPGA firmware download etc is controlled by the driver.

EDIT:
The one I'm looking at is 30% cheaper because its described as 'Grade B: SSD Health between 70-99%'
I imagine that refers to it using up it's overprovisioning and I can assume it won't matter to me?
My use case will be 90% read ie program storage for Solidworks etc with occasional writes by VMs when I'm playing with them. Its not going to be running a database with 10k users or anything. If I went over one or two drive writes a year I'd be astonished.
 
Last edited:
  • Like
Reactions: Samir

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
No, no OPROM for FIO cards.

As for drive health, I'd have to see the status of fio-status -a to confirm anything.

It isn't using up any over-provisioning, as every block on the drive is used/written to every day. It's just used a bit of its wear life, that's all.
 
  • Like
Reactions: YardBouncer

YardBouncer

always yield to the hands-on imperative
Jul 13, 2019
50
27
18
UK
Fantastic, sounds like I don't have to worry about it too much then.
I thought it might be a count of dead NAND cells that had been remapped; Revodrives recommended leaving about 10% unpartitioned for that reason.
The OPROM thing is great news too and wear life won't be an issue for my use case.

The HP Z820 has pretty good thermal design but just in case I shall tape a thermocouple to FIO's heatsink to see how high the temps get when I hammer it for an hour or two with IOMeter or the like.
If they get uncomfortably high I'll add a local fan. Gluing a thermistor to the HS with a small microcontroller PWMing the fan accordingly should stop it screaming all the time.
I know a rise of 10 or 20 degrees science can halve the life of caps and seriously degrade silicon, particularly FPGAs.

I'm so looking forward to having bulk storage with such insane IOPs!
Solidworks comes with various 'toolboxes' that have thousands of standard fasteners and the like and they are mostly small files.
Even with decent quality SATA SSDs it loads the toolbox slowly. Much faster than a rust drive of course but still disappointing for an SSD. In my experience its small file performance at QD1 that defines how subjectively snappy an SSD feels, for workstation use at least. And with 3.2TB available I can dump a copy of my music collection on there too.
Winamp will love it; I'm one of those preverts with multi thousand entry playlists.

Only minor downside I can see is not being able to sleep/hibernate my box any more; I'm sure I read that FIO cards don't like that. Which is fair enough I suppose, its not something servers do.

The price delta between the 'Grade B: SSD Health between 70-99%' and the 'refurbished' ones with no mention of health is about $75.
The 'refurbished' ones are about $300 and the Grade B ones are about $225.
Grade B has 90 days warranty, the others have 1 year.

I shall buy mine tomorrow if possible.
Once I've secured it I'll post the source in the deals section - even though there are about 40 available I want to be sure of snagging one before those with deeper pockets swoop down and buy the lot.

I've no idea if those are good prices or not but they seem pretty good to me.

The other advantages to buying a Grade B one are that it'll be well past it's infant mortality phase and its more likely to have up to date firmware installed.

Many thanks for the information, doubt I'd take the plunge without it.
I'm raising a glass in your general direction.
 

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
"fio-status -a" will give you exact temps of the card, no need for thermocouples/etc.

On any VSL3 or higher card, infant mortality stuff isn't an issue. The design was changed between VSL2 and VSL3 such that dedicated parity chips were no longer a thing and parity was changed to be done at the page level so that bad chunks of nand could be remapped around on the fly.
 
  • Like
Reactions: Samir

YardBouncer

always yield to the hands-on imperative
Jul 13, 2019
50
27
18
UK
The design was changed between VSL2 and VSL3 such that dedicated parity chips were no longer a thing and parity was changed to be done at the page level so that bad chunks of nand could be remapped around on the fly.
That's pretty sweet, I'm increasingly impressed by the design of these things. I suppose decisions like that made it easier to be NAND agnostic.
You gets what you pays for; no wonder they were 12.5k a pop. I'm still astonished by how cheap they are now.
They look nice too and not in a strobing RGB Buck Rogers plastic ray gun kind of way either.

Think I'll just set up automatic polling of fio-status -a and dump the output into a text file I can use to draw pretty graphs.

Since I'm primarily an automation hardware type my immediate thought was to physically instrument the thing, plus I have all the bits on hand.
But writing a .bat or PS script to scrape the output for the values I want is easier for sure.

I do LOVE me some industrial kit. Consumer stuff is cost optimised to the n-th degree and only just good enough to do the job till a month after the warranty runs out. Takes some pretty clever but soul destroying engineering to do that in a repeatable way.
I wouldn't trust the temp sensors in consumer PCs since they mostly just use a diode. Kind of good enough but not what you'd call accurate. But on the other hand diodes are essentially free.

I imagine the FIO cards use a decent I2C sensor or something. Not free but not much out of 12.5k either.

Used to enjoy putting my own PCs together but since discovering surplus workstations I haven't bothered. Loved the Dell Precision line from the 690 to the T7500 but when they changed the case design I jumped ship to a HP z820.
Great machines but you need the wife to hold in your hernia when you pick one up.

Xeons + ECC RAM = life. No more bluescreens ever, it all just works. And since CPU performance has, with rare exceptions, plateaued over recent years I don't really mind that I'm still on a dual E5-2637 v2 box. It's good enough for the girls I go out with.
Skylakes, Platinums and all that can wait till a loaded dual CPU box is 1k on ebay.
 
  • Like
Reactions: Samir

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
I run all E5-26xx v2 xenons at home. Supermicro boards with 2648L CPUs.

As for industrial, yeah, we had to do barometric pressure testing/etc on these, as we had a customer who wanted to put them in a blimp at 20kft as an aerial datacenter. Plenty cold up there, but not much air to pull the heat away. That plus some DoD contracts that had no refrigerated cooling.

The ioDrive2 is pretty solid. Industrial fpga that is good for 100C (we take it offline around 96-98C). The NAND itself doesn't really care much about temps. The hotter the better really (within reason).
 
  • Like
Reactions: Samir

disagreerocket

New Member
Jul 27, 2019
3
0
1
Has anyone been able to resolve: "Status unknown: Driver is in MINIMAL MODE: General channel initialization failure"?

I am unable to perform fio-sure-erase or fio-attach/detach

Code:
[root@ESXi:~] fio-sure-erase /dev/fct0
WARNING: sanitizing will destroy any existing data on the device!
Do you wish to continue [y/n]? y
Erasing blocks: [====================] (100%)
Error: Device '/dev/fct0' is not in a valid state.
   Sure erase requires the device to be detached.
Failure: Fusion-io ioCache 600GB PN:F00-001-600G-CS-0001 SN:1206D4195 FIO-SN:1206D4195 Device:/dev/fct0
Read through this thread several times, and spent probably about 10 hours to get my ioDrive2 running 3.2.16 build 1731 with firmware v7.1.17, rev 116786. I am running ESXi 6.5u2.

Code:
[root@ESXi:~] fio-status -a

Found 1 ioMemory device in this system
Driver version: 3.2.16 build 1731

Adapter: Single Controller Adapter
        Fusion-io ioCache 600GB, Product Number:F00-001-600G-CS-0001, SN:1206D4195, FIO SN:1206D4195
        ioDrive2 Adapter Controller, PN:pA004221001
        External Power: NOT connected
        PCIe Power limit threshold: 24.75W
        PCIe slot available power: unavailable
        Connected ioMemory modules:
          fct0: Product Number:F00-001-600G-CS-0001, SN:1206D4195

fct0    Status unknown: Driver is in MINIMAL MODE:
                General channel initialization failure
        ioDrive2 Adapter Controller, Product Number:F00-001-600G-CS-0001, SN:1206D4195
!! ---> There are active errors or warnings on this device!  Read below for details.
        ioDrive2 Adapter Controller, PN:pA004221001
        SMP(AVR) Versions: App Version: 1.0.35.0, Boot Version: 0.0.9.1
        Located in slot 0 Center of ioDrive2 Adapter Controller SN:1206D4195
        Powerloss protection: not available
        PCI:81:00.0, Slot Number:5
        Vendor:1aed, Device:2001, Sub vendor:1aed, Sub device:2001
        Firmware v7.1.17, rev 116786 Public
        Geometry and capacity information not available.
        Format: not low-level formatted
        PCIe slot available power: 75.00W
        PCIe negotiated link: 4 lanes at 5.0 Gt/sec each, 2000.00 MBytes/sec total
        Internal temperature: 81.70 degC, max 81.70 degC
        Internal voltage: avg 1.02V, max 1.02V
        Aux voltage: avg 2.48V, max 2.48V
        Rated PBW: 8.00 PB
        Lifetime data volumes:
           Physical bytes written: 0
           Physical bytes read   : 0
        RAM usage:
           Current: 0 bytes
           Peak   : 0 bytes

        ACTIVE WARNINGS:
            The ioMemory is currently running in a minimal state.
 

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
Has anyone been able to resolve: "Status unknown: Driver is in MINIMAL MODE: General channel initialization failure"?

I am unable to perform fio-sure-erase or fio-attach/detach

Code:
[root@ESXi:~] fio-sure-erase /dev/fct0
WARNING: sanitizing will destroy any existing data on the device!
Do you wish to continue [y/n]? y
Erasing blocks: [====================] (100%)
Error: Device '/dev/fct0' is not in a valid state.
   Sure erase requires the device to be detached.
Failure: Fusion-io ioCache 600GB PN:F00-001-600G-CS-0001 SN:1206D4195 FIO-SN:1206D4195 Device:/dev/fct0
Read through this thread several times, and spent probably about 10 hours to get my ioDrive2 running 3.2.16 build 1731 with firmware v7.1.17, rev 116786. I am running ESXi 6.5u2.

Code:
[root@ESXi:~] fio-status -a

Found 1 ioMemory device in this system
Driver version: 3.2.16 build 1731

Adapter: Single Controller Adapter
        Fusion-io ioCache 600GB, Product Number:F00-001-600G-CS-0001, SN:1206D4195, FIO SN:1206D4195
        ioDrive2 Adapter Controller, PN:pA004221001
        External Power: NOT connected
        PCIe Power limit threshold: 24.75W
        PCIe slot available power: unavailable
        Connected ioMemory modules:
          fct0: Product Number:F00-001-600G-CS-0001, SN:1206D4195

fct0    Status unknown: Driver is in MINIMAL MODE:
                General channel initialization failure
        ioDrive2 Adapter Controller, Product Number:F00-001-600G-CS-0001, SN:1206D4195
!! ---> There are active errors or warnings on this device!  Read below for details.
        ioDrive2 Adapter Controller, PN:pA004221001
        SMP(AVR) Versions: App Version: 1.0.35.0, Boot Version: 0.0.9.1
        Located in slot 0 Center of ioDrive2 Adapter Controller SN:1206D4195
        Powerloss protection: not available
        PCI:81:00.0, Slot Number:5
        Vendor:1aed, Device:2001, Sub vendor:1aed, Sub device:2001
        Firmware v7.1.17, rev 116786 Public
        Geometry and capacity information not available.
        Format: not low-level formatted
        PCIe slot available power: 75.00W
        PCIe negotiated link: 4 lanes at 5.0 Gt/sec each, 2000.00 MBytes/sec total
        Internal temperature: 81.70 degC, max 81.70 degC
        Internal voltage: avg 1.02V, max 1.02V
        Aux voltage: avg 2.48V, max 2.48V
        Rated PBW: 8.00 PB
        Lifetime data volumes:
           Physical bytes written: 0
           Physical bytes read   : 0
        RAM usage:
           Current: 0 bytes
           Peak   : 0 bytes

        ACTIVE WARNINGS:
            The ioMemory is currently running in a minimal state.
What does fio-pci-check show? I think you need to run it with -a for it to dump all the info?
 

disagreerocket

New Member
Jul 27, 2019
3
0
1
What does fio-pci-check show? I think you need to run it with -a for it to dump all the info?
no -a, but there is a verbose switch:

Code:
[root@ESXi:~] fio-pci-check -v

Root Bridge PCIe 2000 MB/sec needed max

      ioDrive 0000:02:00.0 (2001) Firmware 116786
        Current control settings: 0x2820
          Correctable Error Reporting: disabled
          Non-Fatal Error Reporting: disabled
          Fatal Error Reporting: disabled
          Unsupported Request Reporting: disabled
          Payload size: 256
          Max read size: 512
        Current status: 0x0000
          Correctable Error(s): None
          Non-Fatal Error(s): None
          Fatal Error(s): None
          Unsupported Request(s): None
        link_capabilities: 0x0000f442
          Maximum link speed: 5.0 Gb/s per lane
          Maximum link width: 4 lanes
          Slot Power limit: 25.0W (25000mw)
        Current link_status: 0x00000042
          Link speed: 5.0 Gb/s per lane
          Link width is 4 lanes
        Current link_control: 0x00000000
          Not modifying link enabled state
          Not forcing retrain of link
 

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
Hmm, that seems fine. Only thing I'd think to do is try to load it on the initial 3.0 driver/firmware and see if you can get a sure-erase done. From there I'd follow the upgrade path in the user guide.
 

disagreerocket

New Member
Jul 27, 2019
3
0
1
Never could get it to work, and am in the process of returning it now. Is there a difference between ioCache and ioDrive? The shroud on the top side says ioCache, while the sticker on the back says ioDrive2 with a part number of F00-001-600G-CS-0001.
 

YardBouncer

always yield to the hands-on imperative
Jul 13, 2019
50
27
18
UK
I done got me one!

The 3.2TB card arrived in excellent condition. Seems to be fully working too.
The firmware was almost up to date and it's endurance is more than adequate for my use.

Output from fio-utils seems normal; under 5PB used out of 20PB write endurance if I understand it correctly.
I suspect its very unlikely I'll use up the rest of that.

@acquacow: There is one last thing I'd like to ask.
From reading the pdfs I get the impression that I can no longer hibernate or sleep my workstation.
Is that correct?
It's a bit of a pain if so; once I have my CAD and PLC workspace laid out and a few hundred tabs open(!) I like to avoid rebooting for a while in order to save state.
Is the prohibition for hibernation as well as sleep?

Below is all I had to do to get it working. Files from WD as card is unbranded.
It is in a PCI-e 3 16x slot. Airflow is good, card seems to stabilise at about 56 degrees science. Theres a big old FPGA under that heatsink.

Code:
C:\Program Files\SanDisk\Fusion ioMemory VSL\3.2.15.1699\Driver\fio-utils>fio_pc
i_check
This utility directly probes the pci configuration ports.
There are possible system stability risks in doing this.
Do you wish to continue [y/n]? y

Root Bridge PCIe 18750 MB/sec needed max
Intel(R) Xeon(R)-áE7 v2/Xeon(R) E5 v2/Core i7 DMI2 - 0E00

      Bridge 0000:40:02.00 (41-41) (8086:0e04)
      Intel(R) Xeon(R)+é-áE7 v2/Xeon(R) E5 v2/Core i7 PCI Express Root Port 2a -
 0E04
          Needed 2000 MB/sec Avail 2000 MB/sec
        * Correctable Error(s): Detected
          Clearing Errors

            ioDrive 0000:41:00.0 (2001) Firmware 116786
            Fusion ioMemory VSL Device
                Slot Power limit: 75.0W (75000mw)


C:\Program Files\SanDisk\Fusion ioMemory VSL\3.2.15.1699\Driver\fio-utils>fio_st
atus -V

Found 1 ioMemory device in this system
Driver version: 3.2.15 build 1699

Adapter: Single Controller Adapter
        Fusion-io ioScale 3.20TB, Product Number:F11-002-3T20-CS-0001, SN:1439D1
5B0, FIO SN:1439D15B0
        ioDrive2 Adapter Controller, PN:PA005064001
        External Power: NOT connected
        PCIe Bus voltage: avg 11.60V
        PCIe Bus current: avg 0.92A
        PCIe Bus power: avg 10.62W
        PCIe Power limit threshold: 24.75W
        PCIe slot available power: unavailable
        Connected ioMemory modules:
          fct0: Product Number:F11-002-3T20-CS-0001, SN:1439D15B0

fct0    Attached
        ioDrive2 Adapter Controller, Product Number:F11-002-3T20-CS-0001, SN:143
9D15B0
        ioDrive2 Adapter Controller, PN:PA005064001
        SMP(AVR) Versions: App Version: 1.0.20.0, Boot Version: 1.0.6.1
        Located in slot 0 Center of ioDrive2 Adapter Controller SN:1439D15B0
        Powerloss protection: protected
        PCI:41:00.0, Slot Number:4
        Vendor:1aed, Device:2001, Sub vendor:1aed, Sub device:2001
        Firmware v7.1.17, rev 116786 Public
        3200.00 GBytes device size
        Format: v500, 781250000 sectors of 4096 bytes
        PCIe slot available power: 75.00W
        PCIe negotiated link: 4 lanes at 5.0 Gt/sec each, 2000.00 MBytes/sec tot
al
        Internal temperature: 51.19 degC, max 51.19 degC
        Internal voltage: avg 1.02V, max 1.02V
        Aux voltage: avg 2.48V, max 2.48V
        Reserve space status: Healthy; Reserves: 100.00%, warn at 10.00%
        Active media: 100.00%
        Rated PBW: 20.00 PB, 76.57% remaining
        Lifetime data volumes:
           Physical bytes written: 4,686,697,557,597,560
           Physical bytes read   : 4,522,346,986,201,200
        RAM usage:
           Current: 150,739,328 bytes
           Peak   : 150,739,328 bytes
        Contained VSUs:
          fct0: ID:0, UUID:8e28d28c-c036-4b9a-95d0-455daff7e65b

fct0    State: Online, Type: block device
        ID:0, UUID:8e28d28c-c036-4b9a-95d0-455daff7e65b
        3200.00 GBytes device size
        Format: 781250000 sectors of 4096 bytes


C:\Program Files\SanDisk\Fusion ioMemory VSL\3.2.15.1699\Driver\fio-utils>fio_up
date_iodrive.exe fusion_3.2.11-20150618.fff
WARNING: DO NOT TURN OFF POWER OR RUN ANY IODRIVE UTILITIES WHILE THE FIRMWARE U
PDATE IS IN PROGRESS
  Please wait...this could take a while

Updating: [====================] (100%) -
 fct0 - successfully updated the following:
 Updated the firmware from 7.1.17 rev 116786 to 7.1.17 rev 116786
  Updated SMPCTRL from 1.0.20 to 1.0.21

Please reboot this machine to activate new firmware.
Then after reboot:

Code:
C:\Program Files\SanDisk\Fusion ioMemory VSL\3.2.15.1699\Driver\fio-utils>fio-config -p FIO_EXTERNAL_POWER_OVERRIDE *:75

C:\Program Files\SanDisk\Fusion ioMemory VSL\3.2.15.1699\Driver\fio-utils>fio_st
atus -a

Found 1 ioMemory device in this system
Driver version: 3.2.15 build 1699

Adapter: Single Controller Adapter
        Fusion-io ioScale 3.20TB, Product Number:F11-002-3T20-CS-0001, SN:1439D1
5B0, FIO SN:1439D15B0
        ioDrive2 Adapter Controller, PN:PA005064001
        External Power Override: ON
        External Power: NOT connected
        PCIe Bus voltage: avg 11.61V
        PCIe Bus current: avg 0.91A
        PCIe Bus power: avg 10.69W
        PCIe Power limit threshold: 74.75W
        PCIe slot available power: unavailable
        Connected ioMemory modules:
          fct0: Product Number:F11-002-3T20-CS-0001, SN:1439D15B0

fct0    Attached
        ioDrive2 Adapter Controller, Product Number:F11-002-3T20-CS-0001, SN:143
9D15B0
        ioDrive2 Adapter Controller, PN:PA005064001
        SMP(AVR) Versions: App Version: 1.0.21.0, Boot Version: 1.0.6.1
        Located in slot 0 Center of ioDrive2 Adapter Controller SN:1439D15B0
        Powerloss protection: protected
        PCI:41:00.0, Slot Number:4
        Vendor:1aed, Device:2001, Sub vendor:1aed, Sub device:2001
        Firmware v7.1.17, rev 116786 Public
        3200.00 GBytes device size
        Format: v500, 781250000 sectors of 4096 bytes
        PCIe slot available power: 75.00W
        PCIe negotiated link: 4 lanes at 5.0 Gt/sec each, 2000.00 MBytes/sec tot
al
        Internal temperature: 52.17 degC, max 52.66 degC
        Internal voltage: avg 1.02V, max 1.02V
        Aux voltage: avg 2.48V, max 2.48V
        Reserve space status: Healthy; Reserves: 100.00%, warn at 10.00%
        Active media: 100.00%
        Rated PBW: 20.00 PB, 76.57% remaining
        Lifetime data volumes:
           Physical bytes written: 4,686,697,693,180,304
           Physical bytes read   : 4,522,347,279,742,032
        RAM usage:
           Current: 150,739,328 bytes
           Peak   : 150,739,328 bytes
        Contained VSUs:
          fct0: ID:0, UUID:8e28d28c-c036-4b9a-95d0-455daff7e65b

fct0    State: Online, Type: block device
        ID:0, UUID:8e28d28c-c036-4b9a-95d0-455daff7e65b
        3200.00 GBytes device size
        Format: 781250000 sectors of 4096 bytes
All seems fine; Windows formatted it and I can read & write much faster than SATA (or my Revodrive 3x2; in IOPs and 4k at least).

Many thanks to @acquacow for the help, much appreciated.
Once I've had some kip I'll post the source in the deals section.
 
  • Like
Reactions: Samir

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
I haven't tested hibernate or sleep in a long while. I enabled c-states on my machine and enable c6 states, and it idles at like 75W, so I really don't care about sleeping it. Win10 kept waking it up anyway, so sleep was a bit useless for a while. Pretty sure I had my iodrive installed when I was sleeping my box.
 

YardBouncer

always yield to the hands-on imperative
Jul 13, 2019
50
27
18
UK
I know what you mean, and for short intervals I'll often do the same. But if I'm away for a couple of weeks I like to save state via hibernation as seeing everything again will jog my memory and get me back in the groove a lot faster.

I'll just give it a go and see how it reacts.

The manuals have all kinds of dire warnings about it but at the moment I only have some copied stuff on the iodrive so theres nothing to lose.
I'll report back what happens.
 

Freebsd1976

Active Member
Feb 23, 2018
208
32
28
Today buy one hp brand iomemory 1.6t and upgrade to latest firmware , it is new , and only 140$.
but only one question , card came with 512b format, should I need change it to 4K? I use this card storage file for media and movies on windows server and txt file for machine learning
 
  • Like
Reactions: Samir

acquacow

Well-Known Member
Feb 15, 2017
577
298
63
38
No, changing it to 4k won't do anything. It might decrease some DRAM usage if you are storing lots of small files, but since you are storing normal user-sized files, I doubt it'll make any difference.
 
  • Like
Reactions: Samir